No Major Differences Found between the Effects of
Microwave-Based and Conventional Heat Treatment
Methods on Two Different Liquid Foods
Gábor Géczi1*, Márk Horváth2, Tı́mea Kaszab3, Gonzalo Garnacho Alemany4
1 Department of Environmental Engineering, Institute for Environmental Engineering Systems, Faculty of Mechanical Engineering, Szent István University, Gödöllő,
Hungary, 2 Department of Chemistry and Biochemistry, Institute of Environmental Science, Faculty of Agricultural and Environmental Sciences, Szent István University,
Gödöllő, Hungary, 3 Department of Physics and Control, Faculty of Food Science, Corvinus University of Budapest, Budapest, Hungary, 4 Escola Politècnica Superior,
University of Vic, Vic, Spain
Abstract
Extension of shelf life and preservation of products are both very important for the food industry. However, just as with
other processes, speed and higher manufacturing performance are also beneficial. Although microwave heating is utilized
in a number of industrial processes, there are many unanswered questions about its effects on foods. Here we analyze
whether the effects of microwave heating with continuous flow are equivalent to those of traditional heat transfer methods.
In our study, the effects of heating of liquid foods by conventional and continuous flow microwave heating were studied.
Among other properties, we compared the stability of the liquid foods between the two heat treatments. Our goal was to
determine whether the continuous flow microwave heating and the conventional heating methods have the same effects
on the liquid foods, and, therefore, whether microwave heat treatment can effectively replace conventional heat
treatments. We have compared the colour, separation phenomena of the samples treated by different methods. For milk,
we also monitored the total viable cell count, for orange juice, vitamin C contents in addition to the taste of the product by
sensory analysis. The majority of the results indicate that the circulating coil microwave method used here is equivalent to
the conventional heating method based on thermal conduction and convection. However, some results in the analysis of
the milk samples show clear differences between heat transfer methods. According to our results, the colour parameters
(lightness, red-green and blue-yellow values) of the microwave treated samples differed not only from the untreated
control, but also from the traditional heat treated samples. The differences are visually undetectable, however, they become
evident through analytical measurement with spectrophotometer. This finding suggests that besides thermal effects,
microwave-based food treatment can alter product properties in other ways as well.
Citation: Géczi G, Horváth M, Kaszab T, Alemany GG (2013) No Major Differences Found between the Effects of Microwave-Based and Conventional Heat
Treatment Methods on Two Different Liquid Foods. PLoS ONE 8(1): e53720. doi:10.1371/journal.pone.0053720
Editor: Karl X. Chai, University of Central Florida, United States of America
Received August 27, 2012; Accepted December 5, 2012; Published January 16, 2013
Copyright: ß 2013 Géczi et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was financially supported by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences (http://mta.hu/english/) and by
TÁMOP 4.2.1.B-11/2/KMR-2011-0003 project (http://szie.hu/uj-lendueletet-ad-kutatasnak-kiemelt-tamop-projekt) as a salary supplement for Gábor Géczi.
Furthermore this work was supported with experimental materials (orange juice and milk) by Gramex 2000 Kft. (Veresegyháza, Hungary, http://www.gramex2000.
hu/en/index.html) and Új Mező Kft. (Egyházasdengeleg, Hungary, http://hazitej.net/hazitej/?q = content/el%C3%A9rhet%C5%91s%C3%A9g%C3%BCnk). Last but
not least this work was supported by Livestock Performance Testing Ltd. (Gödöllő, Hungary, http://www.atkft.hu/hu/) for the tests carried out. The funders had no
role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: During the course of this study, the authors have received funding from three commercial sources (Gramex 2000 Kft, Új Mező Kft and
Livestock Performance Testing Ltd.) in the form of access to research materials and/or services without charge. This does not alter the authors9 adherence to all
the PLOS ONE policies on sharing data and materials.
* E-mail: geczi.gabor@gek.szie.hu
The higher speed of the internal heating - compared to that of
the traditional methods based on external heat transfer and heat
conduction - made microwave equipment popular for these
applications. Its thermal application is characterized by release of
thermal energy inside the material being heated. One of the most
significant benefits of microwave heating is reduced damage to
nutritional value due to the shorter duration of the heat treatment
[1], [2], [3], [4]. Additional advantages include energy savings,
lower operating costs, fast processing and flexibility; all of which
could make the use of microwave attractive in both industrial and
small-scale applications [5], [6].
The application of industrial microwave heating using a
microwave-based milk pasteurizer operating in continuous mode
on 82.2uC was tested in the milk industry by Hamid and
Introduction
Microwave heat treatment of food products results in both
thermal and non-thermal effects. The non-thermal effects are
reactions and processes, during which the physical, chemical or
biological conditions of the product change without an increase in
its temperature.
The application of well-known heat transfer mechanisms
(conduction, convection, radiation) and - by another classification
- direct and indirect heating methods offer a number of
possibilities for heat treatments of food products. For heat
treatments of liquid food, the most frequently used methods are
plate- or tube-based heat exchangers, where heat transfer is
applied indirectly using water or steam.
PLOS ONE | www.plosone.org
1
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
[24]. Csapó et al. and Albert et al. developed a continuous mode
microwave method for microwave pasteurization. Their findings
conclude that, in contrast to convection technology, the microwave treatment may result in a higher degree of damage to
vitamin C [25], [26].
We developed in-house instrumentation which enabled a
comparative study of the effects of different heating methods. In
this study we compare the effects of home-made circulating
microwave heating to conventional heating with identical heating
time and temperature with orange juice and milk as a models.
colleagues [7] to decrease the count of those bacteria that are
responsible for spoiling food. Besides applying the microwave heat
treatment successfully in a few areas of the food industry (e.g.
sterilization, dehydration, leavening or baking of bread; for review
see [8], [9]), the heat treatment of liquid foods has also been
attempted. The main problem with liquid heating by microwave is
that during heating by microwave the sterilization effect is not
guaranteed due to uneven temperatures within the product. In
addition, there is a view in the popular press that the use of
microwave energy may have adverse effects. Although there are
few properly conducted studies published in the peer-reviewed
literature to support this view (see e.g. [10], [11], as examples), this
view continues to exert an adverse effect; namely, inhibiting the
spread of technology both in industry and households.
When the vitamin B1, B2 and B6 contents of milk were
analyzed following treatments of microwave and traditional (tube
heat exchanger) heating methods, it was found that neither of the
two methods caused a decrease in these vitamins [3], [4].
Microwave heat treatment was shown to be useful for mild milk
pasteurization [12]. We can also find additional examples for the
analysis of the vitamin content in microwave heat-treated milk.
For example, certain treatment combinations (e.g. 520W, for 4
minutes, ,83uC or 500W, for 6 minutes) caused a higher degree
of decomposition of vitamins A and B12, but in other aspects these
publications also prove the advantages of the microwave and its
applicability for heat treatment [13], [14].
The effects of microwave heat treatment on fruit juices have
been studied at length as this process is often part of their
manufacturing. The quality of citrus is determined by the enzyme
reactions in the fruit not only in the growth phase, but during
processing as well. For example, inactivating the methyl ester
pectin is especially important in prolonging shelf life. Studies
indicate that the pasteurization of fruit juices can be performed
faster with smaller decrease in ascorbic acid at the same time. In
case of citrus juices, the latter clearly determines freshness [2].
Microwave heat treatment, as a feasible alternative in food
processing technology, significantly decreased the initial bacterial
count of fruit [6]. It has also been proven that the decrease of the
amount of mold (Aspergillus sp) is more substantial using
microwave-based heat treatment [15].
Since it was proposed that microwave heat treatment may have
other effects in addition to thermal ones, some researchers have
been focusing on demonstrating these non-thermal effects. A
number of authors have carried out low temperature enzyme
inactivation with microwave irradiation to prove the existence of
the non-thermal effects [16], [17], [18], [19]. However, data from
Shazman suggests that non-thermal events due to microwave
heating do not exist based on their large-scale experiments [20].
They studied the products of the Maillard reaction, the
denaturation of proteins, as well as NaCl solubility after
microwave treatment. They also performed mutagenicity tests.
In contrast with earlier publications, the non-thermal effects did
not occur significantly.
Neményi et al. prove using a variety of liquid foods that
achieving homogeneous heating in an intermittent operating mode
microwave electromagnetic field is difficult. Even heating is
achieved using water traps and the effects of the microwave
radiation on the behavior of the lipase and xanthine oxidase
enzymes in milk are studied [21], [22]. Szerencsi et al. have
established that the spreading of Saccharomyces cerevisiae cells
was greater due to the microwave radiation, which can be
attributed to its non-thermal effects [23]. Lakatos et al. have
detected differences in the size of the milk fat globules treated with
microwave in contrast to traditional pasteurization technologies
PLOS ONE | www.plosone.org
Materials and Methods
Ethics Statement
We have not performed any experiment on animals or humans.
Although volunteers were asked to perform sensory tests, liquid
foods identical or very similar to those that can be purchased from
commercial sources were used in these tests, the general
precautions of food safety were taken into consideration and
volunteers were informed about these facts. Therefore, in
agreement with the corresponding laws of the EU (852/2004/
EK), such tests would not require special permissions.
Orange Juice and Milk Samples
Orange juice samples were freshly squeezed from commercially
available Navelina oranges originating from Seville (Spain)
purchased at local supermarkets. For each test we used 13–15 kg
orange to obtain 5 litres of orange juice. The orange juice was
filtered using a coarse plastic filter of a Hauser brand household
blender to remove the seeds and pulp and was then poured into 5liter plastic jugs. The physical parameters, specifically the total
acid content, vitamin C content, pH and the 24-hour sedimentation of the freshly squeezed orange juice varied in a wider range
during the six-month duration of experiments than those of the
concentrated orange juice samples (Table S1).
We also used non-pasteurized orange juice produced from
concentrate provided by Gramex Ltd. (Veresegyház, Hungary).
This juice required no preparation. The company provided the
material for the experiments in 5-liter plastic bottles obtained from
the production line after mixing, but before heat treatments.
Fresh milk samples were obtained from Új Mező Ltd.
(Egyházasdengeleg, Hungary). After morning milking, filtering
and cooling to 4uC, the samples were transported at ,9uC to the
experiment site. Filtering was performed directly after milking and
before cooling using a paper filter with a 3.25 litre/min/cm2
filtration rate (DeLaval 620660, 60 g/m2, Finland). According to
test reports by Livestock Performance Testing Ltd. of Gödöllő, the
milk samples arrived from the production site with 3.7160.21g/
100 g fat content and 3.3260.14 g/100 g protein content. The
total viable cell count varied between 50.000 and 350.000 CFU/
cm3.
The samples were coded before performing the tests. Doubleblind tests were performed: the person performing the analysis
received mixed and coded samples and the person doing the
treatments did not partake in analyzing the samples.
Reference Measurement Configuration and the Methods
of Heating
The test equipment was implemented by converting a
household microwave oven into a flow-through, continuous
operating mode device with 900 Watts output power. Two holes
with seven mm diameter located 8 cm apart were made in the
oven to introduce and drain the liquid. The microwave equipment
complete with special glass spirals was connected to a STENNER
2
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
Figure 1. Modular experimental setup for heat-treatments of orange juice and milk. Panel A: flow-through heating; Panel B: heating with
recirculation; MH – treated with microwave; TH – traditional heat treatment; WH – untreated control.
doi:10.1371/journal.pone.0053720.g001
temperature measuring system. During each measurement, both
(T1) and (T2) were kept constant. The treated and the control
samples were fed into the storage containers according to the tests
and their mass was measured using the Denver XP-3000 scale.
With the exception of the samples prepared for the ascorbic acid
content test in the case of orange juice, we did not perform holding
– the samples cooled down to the 8 to 10uC storage temperature
naturally. The heating apparatus was supplemented with a threeway valve, which was used to mix the heated sample back to the
storage container (see Panel B in Figure 1). By mixing the sample,
we achieved the target temperature of 85uC (T3) in 13 minutes
with both heating methods. Holding at the set temperature was
achieved by continuing the flowing and heating for another 10
minutes before filling the liquid into the sample storage containers.
During this test, the flow rate was set to Q = 1.19 cm3/s, and the
temperature of the water bath thermostat was kept at 94.5uC
(Figure S2).
85M5 type adjustable feed rate peristaltic pump (Stenner Pump
Company, Jacksonville, FL, USA), an XP-3000 type analytical
scale (Denver Instrument GmbH., Gottingen, Germany) and an
ALMEMO 2590-9 temperature measuring instrument (Ahlborn,
Holzkirchen, Germany) (Panel A in Figure 1).
Inside the microwave oven, the liquid foods flowing through the
glass spirals can be heated to the desired temperature depending
on the length of the spiral and the flow rate of the peristaltic pump.
The temperature can be easily monitored before entering and
after leaving the microwave field, allowing the process to be
controlled effectively. One of the advantages of this method is the
gradual heating and constant output temperature due to the use of
glass spirals. This way the temperature fluctuation characteristics
of intermittent operation can be avoided. The gradual heating and
the resulting temperature differences are presented in Figure S1.
To study the treatments where the milk/orange juice are heated
in different ways, but under identical circumstances (i.e. the final
temperature and the treatment time must be identical), the glass
spiral implement was also used with a T-PHYWE type water bath
(Lauda DR.R. Wobser GmbH, Lauda-Königshofen, Germany).
By adjusting the water temperature, we were able to create the
same treatment temperature as with the microwave method using
an identical flow rate resulting in identical treatment time
(Figure 1). This parallel process made it possible to compare
milk/orange juice samples treated under identical circumstances,
but with different heating methods. As control samples, milk and
orange juice were passed through the glass spiral with no heating.
As seen in Figure 1, the liquid foods were pumped through the the
glass spirals with the peristaltic pump.
For each comparative test, the glass spirals were first placed into
the Whirlpool AT 314 microwave oven (MH) and then into the TPHYWE water bath thermostat (TH), respectively. The same glass
spirals were used for the control samples (WH) without any
heating. The temperature was continuously monitored before (T1)
and after (T2) the treatment using the ALMEMO 2590-9
PLOS ONE | www.plosone.org
Storage Experiments
During the comparative storage experiments for both the milk
and the orange juice we tested conditions that are uncharacteristic
for each particular product. The samples were normally stored in a
dark cellar, where the temperature varied between 8 to 10uC. The
samples were continuously photographed using a Canon Powershot A430 camera. For the test under so-called accelerated
conditions, we poured 20 g properly prepared orange juice into a
Petri dish and stored it on open air at a room temperature of 21–
23uC. The deterioration processes are accelerated when contacting air on a large surface. We repeated the observation four times
with the freshly squeezed orange juice and three times with the
concentrated orange juice.
To observe degradation of orange juice under normal storage
experiments, 100 ml orange juice pre-treated in different ways was
stored in polyethylene terephthalate (PET) bottles. Within this
experiment, samples were divided into three additional treatment
3
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
either the preparation or the experiments and analysis. We applied
a triangle test, where the assessors’ task was to select the distinct
third sample different form the other two [32], [33], [34]. If the
assessor did not detected any difference between the samples, we
asked them not to guess but mark the choice ‘‘CANNOT
DETECT DIFFERENCE’’. We’d like to emphasize that the
participants were assured that one of the three samples is different
from the other two identical ones. The purpose of this test was to
establish if there was any difference between the taste of the
treated and untreated sample and whether or not the two
treatment methods can be distinguished based on the taste of the
samples (Table S2– the arrangement of the triangle test).
Determination of bacterial counts, and quantification of protein
and fat content of the milk samples were performed at the
laboratories of Livestock Performance Testing Ltd. In order to
determine the composition of the milk samples, an FT 6000 series
Milko Scan spectrophotometer (Foss Electric, Denmark) was used.
We created four daily statistical samples from each sample heated
with the two different methods and the control sample (no heating)
as well. The tests were performed a total of 17 times. The initial
parameters of the tests were varied due to the nature of the
procedure (fat content 3.7160.21 g/100 g, protein content
3.3260.14 g/100 g, total viable cell count 50.000–
350.000 CFU/cm3). While creating the sample groups, the target
temperature was kept constant during the day, but was varied
between 64–82uC on each occasion. The target temperature was
controlled by adjusting the flow rate or by replacing the spiral. To
statistically demonstrate the decrease of the bacterial count we
analysed two sample groups with a larger number of samples (16–
16 and 24–24 parallels).
groups: open, not sealed; sealed; sealed and shaken daily. The
storage temperature was 8–10uC. We conducted the test
simultaneously on two sets of nine PET bottles. We repeated the
tests three times with the freshly squeezed orange juice and twice
with the concentrated orange juice. Regarding the changes during
the storage experiments, we based our conclusions on the photos.
Analysis of Food Properties
The comparative tests were performed on the experimental
equipment assembled in the workshop of the Faculty of
Mechanical Engineering at Szent István University. The analysis
and testing of the orange juice and milk samples were conducted at
the Department of Chemistry and Biochemistry, Szent István
University and at the Department of Physics and Control, Faculty
of Food Science, Corvinus University of Budapest. The vitamin C
content and bacterial cell counts were performed by Livestock
Performance Testing Ltd. After the normal storage tests, 12 ml gas
samples were taken from the air above the orange juice by
puncturing the plastic bottles. The gas samples were then injected
into previously evacuated 12 ml volume special sampling tubes
and were shipped to the laboratory to be directly analysed using a
gas chromatograph (HP 5890 Serial II, USA). For the carbondioxide test, we utilized a TCD detector measuring thermal
conductivity.
The colour properties of the milk and orange juice samples were
determined using a ColorLite sph 850 spectrophotometer. Test
results were obtained as CIE (Commission Internationale de la
Éclargie) L*, a*, b* colour properties with wavelengths between
400 and 700 nm [27], [28], [29]. The settings of the instrument
were the following: ‘‘2u standard observer’’ and ‘‘standard
illuminant D65’’. Results of each measurement were calculated
from the average of three measurements by the ColorLite
equipment. There are more than 6 million colour codes in the
CIE Lab System. The colour parameters were the following:
lightness - L*, which defines the grades of brightness from black to
white, red-green colour coordinate - a* and yellow-blue colour
coordinate - b* (Figure S3).
The colour parameters of the foods were monitored every 2–3
days throughout sample storage experiments for the orange juice.
This required a large number of samples, as once the samples were
placed into the spectrophotometer, they could not be used again.
In case of orange juice samples, we prepared 15 aliquots of 25 ml
sealed samples by sample groups (MH, TH and WH). From each
group 3 samples were measured on 1st, 3rd, 6th, 8th and 10th day
following the treatment.
The colour parameters of the milk samples were measured only
on the day following the treatment. The cream forming during the
storage of high-fat milk did not allow continuous, reliable colour
measurement similar to that of the orange juice. We prepared 12
aliquots of 25 ml samples each from the three sample groups,
which were tested on the day following the treatment. The visual
examinations for milk were performed on five different days (on
five different biological samples).
To determine the vitamin C content of the orange juice we
asked the assistance of the aforementioned laboratory of Livestock
Performance Testing Ltd, where the vitamin content of the
beverages were measured using 2,6-dichlorophenolindophenol
reaction with a titration method [30], [31]. We did not participate
directly in these tests. We prepared 12 aliquots of 75 ml statistical
samples from all five sample groups (WH, MH-A, TH-A, MH-B,
TH-B) based on the pairs presented in Figure 1.
The taste tests of the orange juice were performed using a
simplified procedure with the help of untrained volunteers
(students and staff of the university), who did not participate in
PLOS ONE | www.plosone.org
Statistical Analyses
We calculated the statistical average of CO2 content, colour
properties, vitamin C content and total viable cell count
parameters of the samples groups created during the tests. To
compare the averages of samples from the two heat treatment
methods and the non-heated samples, we applied the Student’s
two-sample t-test. We compared the total viable cell count
decreasing effect of the heat treatments of milk, and in case of
orange juice, the effect of the heat treatments on the vitamin C
content.
When evaluating the taste comparison test of orange juice, we
applied the laws of probability theory to decide whether or not the
participants’ answers were given randomly. In some cases, such as
when presenting the colour components or the increase in CO2
content during the storage of orange juice, we used basic
descriptive statistical methods and the results are displayed on
diagrams.
Results
There is no difference between the Beneficial Effects of
Microwave and Traditional Heat Treatment on the Shelf
Life, the Sedimentation of Fractions and the Generation
of CO2 During the Storage of Orange Juice
First, we demonstrate an example of storing orange juice at
room temperature and in open air. There was no visible
deterioration in either sample on the day following the creation
of the samples and placing them in Petri dishes (Figure 2). The
water content of the orange juice evaporated on the relatively large
surface and the fibres dried out. On the untreated samples, intense
moulding was visible even after one week, while on the heattreated samples there was no mould or only traces of mould were
4
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
Figure 2. Both the microwave-based and the traditional heat treatment effectively delays surface deterioration of freshly squeezed
orange juice samples at room temperature. MH – treated with microwave; TH – traditional heat treatment; WH – untreated control.
doi:10.1371/journal.pone.0053720.g002
contained a relatively high amount of CO2 (approx. 1.4 g/L). The
CO2 content of the undisturbed sealed samples was higher by the
order of two magnitudes, while in case of the samples shaken daily
was lower by 40% (Figure 4). The samples treated with different
methods, but stored under the same conditions did not exhibit any
differences.
visible (Figure 2). We did not observe any difference in changes
during storage between the heat treated samples by visual survey.
After evaluating the sedimentation of the orange juice pulp in
test tubes we concluded that the sedimentation of fraction slowed
down compared to the untreated samples, but no differences were
found between the samples treated with the two methods (data not
shown).
Figure 3 shows an example of the changes during storage of
orange juice from concentrate under different conditions. Microwave-treated, traditional heat-treated (in the water bath thermostat) and untreated control samples were poured into PET bottles
and were stored for 34 days in three different ways. The first three
bottles (one from each treatment type) were left open, the next
three were sealed, but left undisturbed, whereas the last three
bottles were also sealed, but were shaken up daily throughout the
whole period.
There were no visible difference in the colour of the samples on
the second day, but we observed that the surface of the samples
that were shaken was foamy (Figure 3, row day 2, VII., VIII., IX.).
On the 8th day the separation of fractions in the first two groups
(I.-VI.) was already visible. Both on days 17 and 34, the untreated
samples (marked ‘29) in any of the three storage methods showed a
difference in colour and foaming compared to the heat treated
samples. The microwave- and water bath-treated samples stored
under the same conditions indicated no visible differences.
Since the sealed bottles used for storing the untreated samples
changed their shape (e.g. Figure 3, day 34, VIII.), samples were
taken from the gas above the juice and were analysed using a gas
chromatograph. We found that the gas above the control samples
PLOS ONE | www.plosone.org
The Two Treatments do not Exert Any Significant Effect
on the Colour, Vitamin C Content or Taste of Orange
Juices
We also determined the colour of the product in CIE Lab
system during the 10 days following the treatment for both the
freshly squeezed orange juice and the orange juice form
concentrate. We could not find any isolated cases across the
sample groups (Figure 5), so neither the heat treatment itself, nor
the method of treatment has a significant effect on the colour of
the product in the early stages of storage. However, the photos of
the previously presented storage experiments (Figure 3) clearly
show that following the onset of deterioration, the untreated
control group (WH) differs from the treated samples in colour as
well.
Besides the colour, the lightness of the product was also
evaluated by determining the so-called L* lightness index.
Although the lightness index of the orange juice from concentrate
showed a slight change with time, the differences were not affected
by the presence/absence or type of heat treatment (Figure 6). The
test performed on freshly squeezed orange juice yielded similar
results (data not shown).
5
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
Figure 3. Storage of orange juice produced from concentrate in open containers, in sealed containers and in sealed containers
shaken daily. Treatment groups of three: First 3 samples: open; second 3 samples: sealed and undisturbed; third 3 samples: sealed, shaken daily.
Markings on the bottles: 1- microwave treated, 2- untreated control, 3- treated with water bath thermostat. The storage test was performed in two
parallel experiments and each gave the results presented in the figure.
doi:10.1371/journal.pone.0053720.g003
the samples from an accredited laboratory. The results clearly
indicated that neither the microwave, nor the traditional heat
treatment method caused significant decrease in vitamin C content
compared to the control group (Figure 7). Statistical analysis
(unequal variance paired t-test) performed on data from 12
replicates from one major batch of milk samples did not indicate
The change in vitamin C content due to heat treatment was also
evaluated. Surprisingly, we found that the vitamin C content had
not decreased during the heat treatment. Therefore, the experiments were repeated with both treatment methods with intense
heating at 85uC followed by holding on temperature. Following
treatment, we requested the analysis of the vitamin C contents of
PLOS ONE | www.plosone.org
6
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
Figure 4. The CO2 content of gas in the bottles decreased similarly in microwave-treated and traditional heat-treated orange juice
samples both in ‘closed’ and ‘closed and shaken’ situations. MH – treated with microwave; TH – traditional heat treatment; WH – untreated
control. Each bar shows the average of six technical replicates.
doi:10.1371/journal.pone.0053720.g004
significant difference in the probability variables (range: 0.06–
0.38), indicating lack of difference between samples treated with
different methods under different conditions (Table 1).
The purpose of the sensory analyses was to find out whether the
participants were able to distinguish on one hand the heat-treated
and untreated orange juice samples and on the other hand the
microwave-heated and water bath thermostat-heated samples
based on taste. The test results are shown in Table 2.
In the test comparing the taste of the treated and untreated
orange juice, most of the participants (MH vs. WH: 37/41; and
TH vs. WH: 38/41) tasted some difference between the heattreated and untreated samples. More than half of these
Figure 5. Colour parameters change of orange juice from concentrate in the period of 10 days following the treatment. The
measured parameters: a* - red-green colour coordinate, b* - blue-yellow colour coordinate. MH – treated with microwave; TH – traditional heat
treatment; WH – untreated control. (The L*-index of samples are shown in Figure 6.).
doi:10.1371/journal.pone.0053720.g005
PLOS ONE | www.plosone.org
7
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
Figure 6. Lightness index (L*) change of orange juice from concentrate in the 10 days following the treatment. MH – treated with
microwave; TH – traditional heat treatment; WH – untreated control. (The a* and b* of samples are shown in Figure 5.).
doi:10.1371/journal.pone.0053720.g006
only 11 persons could find the different sample. When evaluating
the answers in series 3 and 4 together, we can conclude that 10
persons (or 23.2% of the respondents) could not sense any
difference in either test. When considering the 27 test reports
where the respondents clearly marked the different samples, there
were only four persons who recognised the sample that was truly
different. This translates to a hit rate of 14.8%, which is more than
the statistical randomness, but the difference in not convincing
enough.
respondents marked the correct answer in both cases. The
difference was even more pronounced when evaluating the
answers given to the two series together. We found that out of
the 36 respondents who tasted a difference in both series, 14
persons selected the samples correctly, giving a hit rate of 38.9% as
opposed to the 10.9% of statistical randomness.
In tests no. 3 and 4, the samples treated with water bath
thermostat and microwave were compared. More than a quarter
of the respondents did not sense any difference in the taste of the
samples (test series 3 12/43, test series 4 14/43) and in both series
Figure 7. The vitamin C content of orange juice has not changed in microwave-treated or heat-treated samples at 856C. WH:
untreated control; MH-A: flow-through microwave treatment with no temperature holding; TH-A: flow-through traditional heat treatment with no
temperature holding; MH-B: microwave heating of recirculated liquid; TH-B: traditional heating of recirculated liquid.
doi:10.1371/journal.pone.0053720.g007
PLOS ONE | www.plosone.org
8
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
The Two Treatments Affect the Bacterial Count and
Phase separation of Milk Samples in a Similar Manner
Table 1. Statistics for evaluating the decrease in vitamin C
contents [mg/100 ml] in orange juice due to heat treatment.
No.
WH
MH-A
TH-A
MH-B
TH-B
1
53.89
53.81
55.78
53.73
53.73
2
54.31
53.21
54.93
53.73
55.03
3
54.31
53.93
53.73
53.73
54.39
4
54.31
55.29
53.73
52.95
54.39
5
55.03
53.81
53.73
53.73
54.81
6
54.39
54.22
54.93
52.95
54.39
7
55.03
54.06
52.95
54.93
53.89
8
55.39
53.21
53.73
53.73
53.89
9
54.31
55.29
52.95
54.93
53.73
10
53.89
53.21
53.73
52.95
53.73
11
54.39
55.29
55.78
54.93
54.81
12
54.39
55.29
53.73
54.93
53.73
WH
MH-A
TH-A
MH-B
TH-B
Average
54.47
54.22
54.14
53.94
54.21
Max
55.39
55.29
55.78
54.93
55.03
Min
53.89
53.21
52.95
52.95
53.73
Median
54.35
54.00
53.73
53.73
54.14
SD
0.45
0.86
0.98
0.80
0.49
tsz value
0.898
1.057
2.010
1.352
P(T, = t) two-tailed
0.38
0.31
0.06
0.19
tp critical two-tailed
2.11
2.12
2.11
2.07
Result
|tsz|,tp
|tsz|,tp
|tsz|,tp
|tsz|,tp
The fresh milk was heat treated in the 64–82uC range with
microwave energy transfer and water bath thermostat, then its
bacterial count was evaluated by comparing it to that of the
untreated sample. In 17 parallel tests lasting for four months a total
of 85 litres of milk was treated and prepared for measurement. We
provided 140 samples (100 ml each) to determine the total viable
cell count. Both heat treatments resulted in a statistically
significant decrease in the total viable cell count compared to
control (P less than 0.0001, paired t-test): heating the fresh milk to
74.160.2uC (without holding on temperature) resulted in a 74.3%
loss of the total viable cell count (Figure 8). There was no
significant difference between the values obtained with the two
different heating methods (Figure 8 and Figure S4; see also Table
S3 for a representative example of the calculations). The
laboratory tests also proved that the protein and fat content of
the milk was unaffected by the heating (data not shown).
Similarly to the orange juice tests, the milk samples were stored
under various conditions and were photographed continuously.
There was a difference in amount of the separated milk fat, as well
as the amount and colour of whey between the heat-treated and
untreated samples (data not shown). However, similar to the
above, no noticeable differences could be observed between the
samples subjected to the two different heat treatment methods.
The Colour of Microwave-treated Samples Differed not
only from the Untreated Control, but also from
Traditional Heat-treated Ones
The colour of the milk samples was evaluated with the same
procedure used for the orange juice. In case of milk, the colour
changes during storage were not monitored, but the effects of heat
treatment on the colour of the products were analysed. According
to our results, the colour of the microwave treated samples differed
not only from the untreated control, but from the traditional heat
treated samples as well (Figure 9). This experiment was repeated
on milk samples obtained on 5 different days, and we observed the
deviation in the colour and brightness of the samples treated with
the two different methods in four cases. During the repetitions, the
target temperature was varied between 72uC and 83uC. According
to our results, the initial bacterial count, the fat content and the
target temperature did not affect the separation of colour
parameters. For the tested samples the average and the 95%
confidence interval of the L*a*b* values versus handling methods
are demonstrated in Figure 9. There were not significant
differences in L*a* values between the groups of milk samples.
However the b* values show significant difference between the
WH: untreated control; MH-A: flow-through (Q = 1.49 cm3/s) microwave
treatment with no temperature holding; TH-A: flow-through (Q = 1.49 cm3/s)
traditional heat treatment with no temperature holding; MH-B: microwave
heating of recirculated liquid (Q = 1.19 cm3/s); TH-B: traditional heating of
recirculated liquid (Q = 1.19 cm3/s).
t – value of t-test to compare the averages of the two sample groups.
sz(index) – value calculated from the dataset.
p(index) – lookup value for a significance level of p = 0,05.
doi:10.1371/journal.pone.0053720.t001
Based on the test results, we can conclude that the participants
were able to differentiate between the heat treated and untreated
samples by sensory test, but they could not clearly distinguish
between the heat treatment methods. The results presented refer
to the freshly squeezed orange juice, but by performing the test on
the orange juice concentrate products with fewer participants, we
obtained similar results (data not shown).
Table 2 Test results for comparison of taste of heat treated and untreated orange juice.
Test
Number of
participants [no.]
1
41
4
37
19
8
10
2
41
3
38
22
9
7
2*
36*
14*
1–2*
Did not sense any
difference [no.]
Sense some
difference [no.]
Proper marking
[no.]
Improper marking 1
[no.]
Improper marking 2
[no.]
3
43
12
31
11
8
12
4
43
14
29
11
9
9
10*
27*
4*
3–4*
*The two test series evaluated together.
doi:10.1371/journal.pone.0053720.t002
PLOS ONE | www.plosone.org
9
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
Figure 8. The total viable cell count in freshly milked milk significantly decreased after either of the two heat treatment methods.
The average, maximum, minimum and median values of 24 statistical samples are shown per sample group.
doi:10.1371/journal.pone.0053720.g008
groups. Furthermore, the L* and a* values were not significantly
different between the MH and TH groups. Finally, the standard
deviations of the L*, a*, b* values in these groups are bigger than
the values of WH group.
Discussion
The microwave-based heat treatment method produces uneven
heating in the product due to the inhomogeneity of the
electromagnetic field. By continuously transferring liquid foods
through a microwave field, a specific heating can be achieved
depending on the length of the glass spiral and the flow rate of the
metering pump. Identical heating levels and intensity can also be
achieved by transferring the liquid through a water bath instead of
microwave field. Adjusting the temperature of the water bath
affects the results, but using this parallel method microwave energy
transfer and heat treatment based on convective heat transfer can
be compared through the properties of the treated food product.
Orange juice samples treated with microwave and traditional
heat-based methods as well as untreated controls were stored for
the long term under different conditions (open, sealed-undisturbed
and sealed-shaken). Based on the carbon-dioxide content of the air
samples taken from the bottles, both treatment methods slowed
down the fermentation processes in orange juice in comparison to
untreated controls. The deterioration processes and the separation
of the fractions in the undisturbed juices were both delayed as well.
For each of the above listed parameters, no differences were
detected between the effects of the two different heat treatment
methods based on the photos taken during the experiments.
We utilised continuous colour monitoring during the storage of
the orange juice. The Lab colour properties of all three control
groups (WH, TH, MH) changed equally during the first 10 days of
storage. Again, the two different treatment methods did not cause
different effects. Interestingly, no negative effects of the heat
treatment methods were detected when evaluating the vitamin C
content. The vitamin C content decreased neither with the flowthrough heating, nor with heated volume back mixing orange juice
held on temperature for 10 minutes. The target temperature was
85uC in both cases.
Figure 9. The colour parameters of milk were different after
microwave treated vs. traditionally heat-treated samples. The
error bar shows the 95% confidence interval (CI) of L*a*b* values at the
three groups. The measured parameters: a* - red-green colour
coordinate, b* - blue-yellow colour coordinate, L*- lightness index.
The results of 12 statistical samples are shown by sample group.
doi:10.1371/journal.pone.0053720.g009
PLOS ONE | www.plosone.org
10
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
The beneficial effects of the heat treatments were obvious when
analysing the total viable cell count in the milk samples. We
performed heat treatments 17 times at a minimum of 64uC and a
maximum of 82uC without holding on temperature. We found
that the decrease in total viable cell count was identical with both
the microwave heat transfer and the water bath heat treatment. In
our studies, differences of the effects of the heat treatment methods
were detected only in the Lab colour characteristics of the milk
samples, both in terms of the colour coordinates and the brightness
index. The variation amounts to four units on a scale of 100, that is
invisible to the naked eye. However, being able to detect a
difference encourages us to continue our studies. To answer the
question posed in the abstract based on the test conducted, we
consider the microwave heating equivalent, but non-thermal
effects cannot be ruled out.
Our studies might provide proof for food processing facilities
that the whole volume of liquid products can be heated up in a
microwave in a homogeneous manner.
Figure S4 Decrease of the total viable cell count in fresh
milk according to the treatment method. Samples were
evaluated daily throughout a period of four days using four
technical replicates per sample group. The temperature of the
treatment: 08/02/2011–70.560,2uC; 09/02/2011–73.860,3uC;
10/02/2011–64.760,2uC; 11/02/2011–71.560,1uC.
(TIF)
Supporting Information
Acknowledgments
Figure S1 Flow-through microwave equipment with
thermal images demonstrating the gradual heating (on
right) and the temperature difference (on left).
(TIF)
We are grateful to the colleagues at the Livestock Performance Testing Ltd.
(Gödöllő, Hungary) for the tests carried out and their invaluable
professional advice. We’d also like to thank Gramex 2000 Kft (Veresegyháza, Hungary) and Új Mező Kft (Egyházasdengeleg, Hungary) for
providing the orange juice concentrate and the fresh untreated milk
samples for the tests.
Table S1 Selected nutritional values and properties of
the orange juices tested.
(DOCX)
Table S2 Triangle test configuration for comparing the
taste of heat-treated and untreated orange juice.
(DOCX)
Table S3 Decrease in total viable cell count due to heat
treatment in fresh milk.
(DOCX)
The temperature measured in the container
for the mix-back heating method versus elapsed time.
MH – treated with microwave; TH – traditional heat treatment;
WH – untreated control.
(TIF)
Figure S2
Author Contributions
Developed the microwave and water bath termostat comparative
experiments: GG. Conceived and designed the experiments: GG MH
TK GGA. Performed the experiments: GG MH TK GGA. Analyzed the
data: GG MH TK GGA. Contributed reagents/materials/analysis tools:
GG MH TK. Wrote the paper: GG TK.
Figure S3 Interpretation of the CIELab system colour
properties [29].
(TIF)
References
13. Sieber R, Eberhard P, Fuchs D, Gallmann PU, Strahm W (1996) Effect of
microwave heating on vitamins A, E, B1, B2 and B6 in milk. Journal of Dairy
Research 63: 169–172.
14. Watanabe F, Abe K, Fujita T, Goto M, Hiemori M, et al. (1998) Effects of
microwave heating on the loss of vitamin B12 in foods. Journal of Agricultural
and Food Chemistry 46: 206–210.
15. Valderrama ÁM, Sánchez RL (2008) Use of microwaves in the treatment of
mango juice (in Spanish). Lasallista de Investigación 5(2): 13–19.
16. de la Hoz A, Diaz-Ortiz A, Moreno A (2005) Microwaves in organic synthesis.
Thermal and non-thermal microwave effects. Chemical Society Reviews 34:
164–178.
17. Porcelli M, Cacciapuoti G, Fusco S, Massa R, d’Ambrosio G et al. (1997) Nontermal effects of microwaves on proteins: thermophilic enzymes as model system.
FEBS Letter 402(2–3): 102–106. doi: 10.1016/S0014–5793(96)01505-0.
18. Banik S, Bandyopadhyay S, Ganguly S (2003) Bioeffects of microwave – a brief
review. Bioresource Technology 87(2): 155–159. doi: 10.1016/S0960–
8524(02)00169-4
19. George DF, Bilek MM, McKenzie DR (2008) Non-Thermal effects in the
microwave induced unfolding of proteins observed by chaperone binding.
Bioeletromagnetics 29(4): 324–330. doi: 10.1002/bem.20382.
20. Shazman A, Mizrahi,S, Cogan U, Shimoni E (2007) Examining for possible
non-thermal effects during heating in a microwave oven. Food Chemistry 103(2):
444–453.
21. Lakatos E (2006): Treatment of liquid foods, particularly the effect of microwave
to milk (in Hungarian).PhD thesis, University of West Hungary, Mosonmagyaróvár, Hungary, 169 p.
22. Neményi M, Lakatos E, Kovács AJ (2006) Examination of milk fat globule
changes in homogeneous microwave field. Journal of Food Physics. 17–18: 29–
42.
23. Szerencsi Á, Lakatos E, Kovács AJ, Neményi M (2009) Non-thermal effect of
microwave treatment on enzyme suspensions Part I.: Water electrolysis.
Analecta Technica Szegedinensia. Norma Nyomdász Kft. Kiadó és Nyomda,
Szeged, Hungary, 58–62.
24. Lakatos E, Kovács AJ, Végváry Gy, Neményi M (2010) The effect of microwave
to the operation of lipase and xanthine oxidase enzymes in consumer milk (in
Hungarian) Magyar Állatorvosok Lapja 132: 728–734.
1. Vadivambal R, Jayas DS (2007) Changes in quality of microwave-treated
agricultural products – a review. BiosystemsEngineering 98(1): 1–16. doi:
10.1016/j.biosystemseng.2007.06.006.
2. Camacho MM; Garcı́a E; Garcı́a M, Martinez N (2009) Microwave
Pasteurization of Grapefruit Juice (in Spanish) Alimentación Equipos y
Tecnologı́a 244: 34–38.
3. Sierra I, Vidal-Valverde C (2000) Influence of heating conditions in continuousflow microwave or tubular heat exchange systems on the vitamin B1 and B2
content of milk. Journal Lait, 80(6): 601–608.
4. Sierra I, Vidal-Valverde C (2001) Vitamin B1 and B6 retention in milk after
continuous flow microwave and conventional heating at high temperatures.
Journal of Food Protection. 64(6): 890–894.
5. Swain MVL, Russell SL, Clarke RN, Swain MJ (2004) The development of food
simulants for microwave oven testing. International Journal of Food Science &
Technology 39: 623–630. doi: 10.1111/j.1365–2621.2004.00819.x.
6. Peremanyer M, Grèbol N (2010) Microwaves in the food industry: Pasteurization of ready-to-serve prepared meals (in Spanish). Alimentació n
Equipos y Tecnologı́a 255: 12–14.
7. Hamid MAK, Boulanger RJ, Tong SC, Gallop RA, Pereira RR (1969)
Microwave pasteurization of raw milk. Journal of Microwave Power. 4(4): 272–
275.
8. Decareau RV (1985) Microwaves in the Food Processing Industry. Academic
Press Inc. Orlando, Florida. 234p.
9. Kingston HM, Haswell SJ (1997) Microwave-Enhanced Chemistry. Fundamentals, Sample Preparation, and Applications. American Chemical Society,
Washinton, DC. 693p.
10. Wayne A, Newel L (2000) The Hidden Hazards Of Microwave Cooking.
Available: http://www.vsan.org/rok-az/misc/HazardsOfMicrowaveCooking.
pdf Accessed: 08.08.2012.
11. Wild M (?) Are Microwave Ovens a Source of Danger? Available: http://
curezone.com/art/read.asp?ID = 112&db = 7&C0 = 1 Accessed: 08.08.2012.
12. Villamiel M, López-Fandino R, Corzo N, Martinez-Castro I, Olano A (1996)
Effects of continous flow microwave treatment on chemical and microbiological
characteristics of milk. Zeitschrift für Lebensmitteluntersuchung und Forschung,
202(1): 15–18.
PLOS ONE | www.plosone.org
11
January 2013 | Volume 8 | Issue 1 | e53720
Effects of Microwave Heating on Liquid Foods
25. Albert Cs, Lányi Sz, Csapóné KZs, Salamon Sz, Csapó J (2008) The effect of
microwave pasteurisation to the composition of milk II. Vitamin B1, B2, B6, B12
and C – useful lysine, lysine-alanin, hydroxymethyl-furfural content (in
Hungarian) Acta Agraria Kaposváriensis 12(3): 25–36.
26. Csapó J, Albert Cs, Lányi Sz, Salamon Sz, Csapóné KZs (2008) The effect of
microwave pasteurisation to the composition of milk I. Amino acid Content, free
amino acid content, biological value (in Hungarian) Acta Agraria Kaposváriensis, 12(3): 11–24.
27. Kaszab T, Kovács Z, Szöllősi D, Fekete A (2011) Prediction of carrot sensory
attributes by mechanical test and electronic tongue. Acta Alimentaria,
40(Suppl.): 41–58 DOI: 10.1556/AAlim.40.2011.Suppl.5.
28. Kaszab T, Kovács Z, Fekete A (2010) Effect of storage to certain quality
parameters of carrot (in Hungarian) Élelmiszer Tudomány Technológia 64(1):
23–24.
29. Pirate M (2010) Color Calibration is the Key to Consistency; the substance of
things hoped for and the evidence of things seen! Available: http://
PLOS ONE | www.plosone.org
30.
31.
32.
33.
34.
12
digitalprintingevolution.blogspot.hu/2010_04_01_archive.html Accessed:
08.08.2012.
Smoot JM, Nagy S (1980) Effects of storage temperature and duration on total
vitamin C content of canned single-strength grapefruit juice. J. Agric. Food
Chem. 28(2): 417–421. doi: 10–1021/jf60228a050.
Haddad P (1977) Vitamin C content of commercial orange juices. An analytical
project. J. Chem. Educ 54(3): 192. doi: 10.1021/ed054p192.
Garcı́a AF, Butz P, Bognár A, Tauscher B (2001) Antioxidative capacity,
nutrient content and sensory quality of orange juice and an orange-lemon-carrot
juice product after high pressure treatment and storage in different packaging
Eur Food Res Technol 213: 290–296 DOI 10.1007/s002170100332.
Morisson DG (1981) Triangle taste tests: are the subjects who respond correctly
lucky or good? Journal of Marketing 45(3): 111–119 DOI 10.2307/1251546.
Stillman JA (1993) Response selection, sensitivity, and taste-test performance
Perception & Psychophysics, 54 (2): 190–194 DOI: 10.3758/BF03211755.
January 2013 | Volume 8 | Issue 1 | e53720